# Creating an Agent with AWS Bedrock

**Video walkthrough** [**here**](https://airiskinc.sharepoint.com/:v:/s/ArtificialIntelligenceRisk/EbN_lrwO8lJFgX-yJDUsD7gBYimugqGDTX3NQOt3fWw1Ew?e=avyzEu\&nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZy1MaW5rIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXcifX0%3D)**.**

**1. Go to the Admin Panel**\
Navigate to the admin section of your dashboard to access agent management tools.

**2. Click “Add New Agent”**\
Locate and select the **“Add New Agent”** button to begin creating a new agent profile.

<figure><img src="/files/ijQx8leyjiT32U63hxe0" alt=""><figcaption><p>Sample inputs for step 3</p></figcaption></figure>

**3. Enter the Agent Name, Description, and Default Prompt. See** [**General Settings**](/admin/ai-agents/adding-a-new-agent/general-settings.md)**.**

<figure><img src="/files/iqtowVovNCQc4tuwE9rv" alt=""><figcaption><p>Sample suggestion input</p></figcaption></figure>

**4. (Optional) Add a Suggested Question**\
In the **Suggested Questions** field, click on "Add Suggestions". Then, add a greeting and description, click the checkmark, and save. You may add as many suggestions as you'd like.

<figure><img src="/files/rZdCxInaGGhgreGV3IeM" alt=""><figcaption></figcaption></figure>

**5. Select the Agent Engine: AWS Bedrock**\
Before saving, make sure to select **AWS Bedrock** as the agent engine.

<figure><img src="/files/bhDtOFmFxKKphSUWiDy5" alt=""><figcaption><p>Sample option inputs for steps 6-8</p></figcaption></figure>

**6. Choose the Model Under AWS Bedrock Settings**\
Under **AWS Bedrock Settings**, choose the appropriate model you want the agent to use.

**7. Check the “Enable File Search” Box**\
To allow your agent to retrieve information from uploaded documents, make sure to check the **“Enable File Search”** box.

**8. Choose an Existing Bucket or Create a New One**\
Type in a bucket name to use either an existing one or create a new one.&#x20;

***Bucket*****:** is a place to store files. It’s safe for different agents to share the same bucket name, as they won’t be able to access or read each other’s files. Users can use an existing bucket or create a new bucket. The buckets are accessible in the AWS S3 Console.

<figure><img src="/files/3E73IjIelVx44Ys0h7jy" alt=""><figcaption><p>Step 9</p></figcaption></figure>

**9. (Optional) Check the “User Can Upload Files” Box**\
If you want end users to be able to upload their own documents via the chat screen for the agent to use, check the **“User Can Upload Files”** box.

<figure><img src="/files/vL1YrpICrdJj00Nfpt99" alt=""><figcaption><p>Step 10</p></figcaption></figure>

**10. Click “Add Files” to Upload Documents for the Agent**\
Use the **“Add Files”** button to upload any documents you want the agent to access. All files assigned to that agent will be accessible across all conversations for a given user.

<figure><img src="/files/5erSk4y8QVnrhNoh352p" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/9G012uJtOvF3WzH6oZEh" alt=""><figcaption><p>Step 11</p></figcaption></figure>

**11. Add Any Data Functions You Want the Agent to Use**\
Scroll to the **Add** **Data** **Functions** section and select any pre-configured functions you’d like this agent to access. These can include things like Web Functions, Custom API Functions, and Custom Database Functions.

<figure><img src="/files/G71P2lHg4I0xtrbgKdpW" alt=""><figcaption><p>Step 12</p></figcaption></figure>

**12. See** [**Advanced Settings**](/admin/ai-agents/adding-a-new-agent/advanced-settings.md)**.**


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.aicrisk.com/admin/ai-agents/adding-a-new-agent/agent-engine-properties/aws-bedrock/creating-an-agent-with-aws-bedrock.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.